Learning Rate Schedules
Linear Warmup With Cosine Annealing
Linear Warmup With Cosine Annealing is a learning rate schedule where we increase the learning rate linearly for $n$ updates and then anneal according to a cosine schedule afterwards.
Papers
| Paper | Code | Results | Date | Stars |
|---|
Tasks
| Task | Papers | Share |
|---|---|---|
| Language Modelling | 67 | 9.25% |
| Large Language Model | 42 | 5.80% |
| Question Answering | 37 | 5.11% |
| Retrieval | 29 | 4.01% |
| In-Context Learning | 25 | 3.45% |
| Text Generation | 23 | 3.18% |
| Sentence | 20 | 2.76% |
| Prompt Engineering | 19 | 2.62% |
| Code Generation | 18 | 2.49% |
Cosine Annealing
